Nickel(II) increases the sensitivity of V79 Chinese hamster cells ...

Carcinogenesis vol.20 no.7 pp.1177–1184, 1999

Nickel(II) increases the sensitivity of V79 Chinese hamster cells towards cisplatin and transplatin by interference with distinct steps of DNA repair

Ines Krueger, Leon H.F.Mullenders1 and Andrea Hartwig2 University of Karlsruhe, Department of Food Chemistry, Postfach 6980, D-76128 Karlsruhe, Germany and 1MGC–Department of Radiation Genetics and Chemical Mutagenesis, Leiden University, Leiden, The Netherlands 2To whom correspondence should be addressed. Email: [email protected]

Nickel compounds are carcinogenic to humans and to experimental animals. In contrast to their weak mutagenicity, they have been shown previously to increase UVinduced cytotoxicity and mutagenicity and to interfere with the repair of UV-induced DNA lesions by disrupting DNA– protein interactions involved in DNA damage recognition. In the present study we applied cisplatin, transplatin and mitomycin C to investigate whether these enhancing effects and DNA repair inhibition are also relevant for other DNA damaging agents. Nickel(II) at non-cytotoxic concentrations of 50 µM and higher caused a pronounced increase in cisplatin-, transplatin- and mitomycin C-induced cytotoxicity, which was neither due to an altered uptake of cisplatin or transplatin nor to an increase in DNA adduct formation. However, nickel(II) inhibited the repair of cisplatin- and transplatin-induced DNA lesions. In combination with transplatin, it decreased the incision frequency, indicating that the DNA damage recognition/incision step during nucleotide excision repair is affected in general by nickel(II). In support of this, concentrations as low as 10 µM nickel(II) decreased binding of the xeroderma pigmentosum complementation group A protein to a cisplatin-damaged oligonucleotide. When combined with cisplatin, the incision frequency was affected only marginally, while nickel(II) led to a marked accumulation of DNA strand breaks, indicating an inhibition of the polymerization/ligation step of the repair process. This effect may be explained by interference with the repair of DNA–DNA interstrand crosslinks induced by cisplatin. Our results suggest that nickel(II) at non-cytotoxic concentrations inhibits nucleotide excision repair and possibly crosslink repair by interference with distinct steps of the respective repair pathways. Introduction Epidemiological studies identified nickel as carcinogenic to occupationally exposed humans, increasing the incidence of lung and nasal cancer. In experimental animals, compounds of nickel induced the formation of lung cancer and local sarcomas in different species. However, the mechanism of cancer induction is not clear. Nickel(II) is not mutagenic in bacterial test systems and in mammalian cells in culture the mutagenic potential is rather weak (1). In contrast to its weak Abbreviations: MEMα, α-modified minimal essential medium; NER, nucleotide excision repair; XPA, xeroderma pigmentosum complementation group A. © Oxford University Press

mutagenic effects, nickel(II) has been shown to enhance cytotoxicity, mutagenicity and sister chromatid exchange induction in V79 Chinese hamster cells in combination with UV light in a more pronounced manner (2). In addition, nickel(II) inhibits the repair of UV-induced DNA lesions by interference with the incision and ligation steps of the nucleotide excision repair (NER) system, which is presumably due to the disruption of DNA–protein interactions involved in DNA damage recognition (3–7). To find out whether these enhancing effects and repair inhibition are restricted to UV-induced DNA damage or also apply to other DNA damaging agents, we elucidated the effect of nickel(II) on the cellular response of V79 Chinese hamster cells treated with cis-diamminedichloroplatinum(II) (cisplatin), trans-diamminedichloroplatinum(II) (transplatin) or mitomycin C. Cisplatin is widely used as an antitumour agent and its cytostatic activity is closely related to the formation and repair of different types of DNA adducts. It forms primarily 1,2intrastrand crosslinks between adjacent purines [d(GpG) and d(ApG)] and to a lesser extent DNA monoadducts, DNA– DNA interstrand crosslinks and DNA–protein crosslinks. While DNA monoadducts, 1,3-intrastrand crosslinks, DNA–DNA interstrand crosslinks and DNA–protein crosslinks are also induced by the therapeutically inactive stereoisomer transplatin, no 1,2-intrastrand crosslinks are formed by this compound and the lower cytotoxicity and therapeutic inactivity of transplatin is thought to be due to the different spectrum of DNA adducts induced by this agent (8,9). One aspect which seems to play a decisive role in drug tolerance is the removal of platinuminduced DNA adducts by cellular repair systems. The main mechanism involved in the repair of bifunctional platinum– DNA lesions is NER. However, the most frequent DNA lesions induced by cisplatin, 1,2-intrastrand crosslinks, are poor substrates for the mammalian NER pathway, whereas transplatin-induced DNA lesions are removed efficiently (10–12). The importance of DNA repair for cell survival after cisplatin treatment has been further demonstrated by the enhanced sensitivity of repair-deficient cell lines. These include xeroderma pigmentosum cells defective in NER and also Fanconi’s anemia cells, which are particularly sensitive to agents capable of inducing DNA–DNA interstrand crosslinks, indicating that this type of lesion also contributes to the cytotoxicity of cisplatin (13–15). Mitomycin C is a chemotherapeutic agent that reacts with the N2 position of guanines forming DNA monoadducts as well as CpG interstrand and GpG intrastrand crosslinks (reviewed in ref. 16). The cytotoxicity of mitomycin C is greatly enhanced in Fanconi’s anemia cells (14,17,18), suggesting that the extent of unrepaired DNA–DNA interstrand crosslinks contributes substantially to cell death. To find out whether nickel(II) interferes with the repair of DNA lesions induced by cisplatin, transplatin and mitomycin C we investigated its effect on the cytotoxicity of these agents and on different steps of DNA repair after treatment with 1177

I.Krueger, L.H.F.Mullenders and A.Hartwig

cisplatin and transplatin. Finally, we determined binding of the xeroderma pigmentosum complementation group A (XPA) protein to cisplatin-damaged DNA in the absence and presence of nickel(II). The XPA protein is a damage recognition protein which binds with high affinity to UV-induced DNA photoproducts and DNA adducts formed by cisplatin (19–22) and which is essential to form the pre-incision complex of the NER system in mammalian cells. We demonstrate that nickel(II) increases the cytotoxicity of all three agents by interfering with distinct steps of repair processes involved in the removal of cisplatin- and transplatin-induced DNA lesions. Materials and methods Materials α-Modified minimal essential medium (MEMα), fetal calf serum and the trypsin and penicillin/streptomycin solutions are products of Gibco (Karlsruhe, Germany). Cisplatin, transplatin, mitomycin C and hydroxyurea were obtained from Sigma Chemical Co. (Munich, Germany). SDS and hydroxyapatite were from Calbiochem (Bad Soden, Germany). The digoxygenin nucleic acid detection kit was from Boehringer (Mannheim, Germany). The digoxygeninlabelled oligonucleotide was obtained from MWG-Biotech (Ebersberg, Germany). The XPA protein was a gift from Dr A.Eker (Rotterdam, The Netherlands). All other chemicals, including the inhibitor aphidicolin, NiCl2·6H2O and Giemsa stain were bought from Merck (Darmstadt, Germany). The culture dishes were supplied by Nunc (Wiesbaden Germany). Cell culture and preparation of chemicals V79 cells were grown as monolayers in MEMα containing 10% fetal calf serum, 100 U/ml penicillin and 100 µg/ml streptomycin. The cells were incubated with 5% CO2 in air and 100% humidification. Stock solutions of nickel and mitomycin C were prepared in bidistilled water and stored at 4°C, while transplatin and cisplatin were freshly dissolved in MEMα without serum before use. Colony forming ability Logarithmically growing V79 cells were treated according to the experimental protocol. Thereafter, they were trypsinized and 300 cells/60 mm dish were seeded in triplicate for determination of colony-forming ability. After 5 days incubation, colonies were fixed with ethanol, stained with Giemsa, counted and calculated as a percentage of the control. Atomic absorption spectrometry V79 cells were pre-incubated with NiCl2 for 20 h and treated with cisplatin or transplatin for 4 h. At the end of the treatment, the cells were washed three times with ice-cold MEMα, trypsinized, counted and mineralized with 65% (v/v) HNO3 and 30% H2O2 (1:1). Uptake was measured with a Perkin Elmer 2380 atomic absorption spectrophotometer equipped with an HGA 400 graphite furnace following the instructions in the manufacturer’s manual. Calculation of the intracellular platinum concentration was done as described previously, based on a V79 cell volume of 1.55310–12 l (23). Alkaline unwinding DNA strand breaks were determined according to the method of Ahnstro¨m and Erixson (24) with modifications as described previously (25). Briefly, V79 cells were treated according to the experimental protocol. Afterwards, the medium was removed and an alkaline solution (0.03 M NaOH, 0.01 M Na2HPO4, 0.9 M NaCl, pH 12.3) was added and incubated in the dark for 30 min. Thereafter, the solution was neutralized with HCl, sonicated and SDS was added yielding a final concentration of 0.05%. Separation of single- and double-stranded DNA was performed on 1 ml high resolution hydroxyapatite columns (Calbiochem) at 60°C. Single- and double-stranded DNA were eluted with 3 ml 0.15 M and 0.35 M potassium phosphate buffer, respectively. The DNA content of both fractions was determined by adding Hoechst 33258 to a final concentration of 7.5310–7 M to 1 ml of each sample and by measuring the fluorescence with a spectrophotometer (LS 50 B; Perkin Elmer) at an excitation wavelength of 360 nm and an emission wavelength of 450 nm. The fraction of double-stranded DNA was calculated as described previously. In order to quantitate the lesion frequencies, the fraction of double-stranded DNA was compared with the amount of double-stranded DNA produced by X-rays in a calibration experiment (26). Gel mobility shift assay The gel mobility shift assay was performed as described by Garabedian et al. (27) with modifications. The DNA probe was 59-end-labelled with digoxygenin and consisted of the following sequence:

1178

Fig. 1. Colony-forming ability after 36 h incubation of V79 cells with nickel(II). Logarithmically growing cells were treated with NiCl2, trypsinized and the colony-forming ability was determined as described in Materials and methods. Shown are mean values 6 SD from three determinations. 59-CATTGTTTAT TAACATTACA ATTACAATTG GTAACACAAT GTTACAAATG TTTAACAATA-39 39-GTAACAAATA ATTGTAATGT TAATGTTAAC CATTGTGTTA CAATGTTTAC AAATTGTTAT-59. The potential cisplatin binding sites are underlined. The oligonucleotide was incubated in the dark with 4 mM cisplatin at 37°C for 1 h. Thereafter, the oligonucleotides were precipitated with ethanol and washed twice. The XPA protein was diluted in buffer A (40 mM HEPES, 20% glycerol, 2.7 mM EDTA, 300 µg/ml bovine serum albumin, 140 mM KCl, 16.13 mM MgCl2 and 1 mM DTT, pH 7.9) to a final concentration of 250 ng/µl. The DNAbinding reaction mixture contained buffer A, 200 fmol oligonucleotide 59end-labelled with digoxygenin and the XPA protein. The XPA protein was pre-incubated with NiCl2 at room temperature for 15 min, whereafter the oligonucleotide was added and incubation was continued for another 30 min. The DNA-binding reaction was stopped by adding buffer B (25 mM Tris, 25 mM borate, 0.5 mM EDTA, 40% glycerol, pH 8.0). The samples were separated electrophoretically in a 4% native polyacrylamide gel in 0.53 TBE buffer (50 mM Tris, 50 mM borate, 1 mM EDTA, pH 8.0) at 90 V for 90 min. Southern blotting was done in a semi-dry electroblotting apparatus applying a positively charged nylon membrane (Hybond-N1; Amersham, Braunschweig, Germany). Detection of the digoxygenin-labelled oligonucleotide was performed colourimetrically with alkaline phosphatase conjugated to an antidigoxygenin antibody using nitroblue tetrazolium salt and 5-bromo-4-chloro3-indoyl phosphate as substrates (Boehringer, Mannheim, Germany). To quantitate the DNA-binding activity of the XPA protein, the respective bands were analysed densitometrically (Kem-En Tec Sofware Systems, Copenhagen, Denmark).

Results Effect of nickel(II) on the cytotoxicity of cisplatin, transplatin and mitomycin C To investigate the effect of nickel(II) on the cellular response to cisplatin, transplatin and mitomycin C, incubation conditions were defined which ensure sufficient uptake of the metal compound and the presence of nickel(II) during a considerable part of the DNA repair process. Nickel(II) is taken up comparatively slowly into cultured mammalian cells, reaching an equilibrium after ~16–24 h incubation (6). Therefore, V79 Chinese hamster cells were pre-incubated with nickel(II) for 20 h, treated with cisplatin, transplatin or mitomycin C for 2 h and post-incubated with nickel(II) for up to 20 h. The cytotoxicity of nickel(II) alone after 36 h incubation is shown in Figure 1. While concentrations up to 250 µM were not cytotoxic, higher concentrations led to a marked decrease in colony-forming ability. Therefore, non-cytotoxic concentra-

Ni(II) and cisplatin and transplatin sensitivity

Fig. 2. Colony-forming ability after combined treatment of V79 cells with nickel(II) and cisplatin. Logarithmically growing cells were pre-incubated with NiCl2 for 20 h, treated with cisplatin in the absence of nickel(II) for 2 h and post-incubated with NiCl2 for 20 h. (A) 250 µM nickel(II) and cisplatin concentrations as indicated; (B) 2.5 µM cisplatin and nickel(II) concentrations as indicated. Shown are mean values 6 SD from three determinations.

Fig. 3. Colony-forming ability of V79 cells after combined treatment with nickel(II) and transplatin. The cells were pre-incubated with 250 µM NiCl2 for 20 h, treated with transplatin for 2 h in the absence of nickel(II) and post-incubated with NiCl2 for 20 h. Shown are mean values 6 SD from three determinations.

Fig. 4. Colony-forming ability of V79 cells after combined treatment with nickel(II) and mitomycin C. The cells were pre-incubated with 250 µM NiCl2 for 20 h, treated with mitomycin C for 1 h in the absence of nickel(II) and post-incubated with NiCl2 for 20 h. Shown are mean values 6 SD from three determinations.

tions up to 250 µM NiCl2 were used in the subsequent experiments. The effect of nickel(II) on the cytotoxicity of cisplatin is indicated in Figure 2. In the absence of nickel(II), V79 cells were able to tolerate concentrations of cisplatin up to 2.5 µM, whereas higher concentrations markedly reduced the colonyforming ability. In contrast, the combined treatment of V79 cells with cisplatin and 250 µM NiCl2 caused a considerable enhancement of cytotoxicity towards the platinum compound; in this case, cisplatin concentrations as low as 1.5 µM decreased the colony-forming ability of V79 cells (Figure 2A). As shown in Figure 2B, this effect is dose dependent with respect to nickel(II); concentrations as low as 50 µM NiCl2 caused a marked reduction in colony-forming ability after cisplatin treatment. In a next step, we investigated the effect of nickel(II) on the cytotoxicity of transplatin. As shown in Figure 3, transplatin is less toxic as compared with cisplatin and concentrations up to 150 µM caused no reduction in colony-forming ability. However, the combined treatment with the non-cytotoxic concentration of 250 µM NiCl2 resulted in a marked increase in transplatin-induced cytotoxicity.

One type of lesion which presumably accounts for the cytotoxicity of cisplatin and, if not repaired, might also contribute to the cytotoxicity of transplatin, is the DNA–DNA interstrand crosslink (28). To investigate whether nickel(II) possibly affects the fate of this DNA adduct, the effect of nickel(II) on the cytotoxicity of mitomycin C was determined. This compound has been chosen since it is believed that the extent of DNA–DNA-interstrand crosslinks contributes largely to its cytotoxcity. Applying the respective incubation conditions as described for the preceding experiments, nickel(II) enhanced the cytotoxicity in combination with this cytostatic agent as well (Figure 4). Effect of nickel(II) on platinum uptake and DNA adduct formation As described above, there is a close relationship between the extent of DNA damage and the cytotoxicity of cisplatin, transplatin and mitomycin C. Therefore, in principle two different reasons could account for the observed increase in platinum- and mitomycin C-induced cytotoxicity in the presence of nickel(II). One reason could be an increase in the initial lesion frequencies, for example due to an increased drug uptake mediated by nickel(II). Alternatively, the cellular 1179


Fig. 5. Effect of nickel(II) on the uptake of cisplatin (A) or transplatin (B) into V79 cells. The cells were incubated with up to 150 µM cisplatin or transplatin for 4 h and the intracellular platinum content was determined by atomic absorption spectrometry. When investigating the effect of nickel(II), the cells were pre-incubated with 250 µM NiCl2 for 20 h before cisplatin or transplatin treatment. Shown are mean values 6 SD from three determinations.

response, including the repair capacity, could be diminished. With respect to the first possibility, we investigated the effect of nickel(II) on the uptake of cisplatin. As shown in Figure 5A, there was a dose-dependent uptake of cisplatin yielding a 2-fold accumulation of intracellular platinum after 4 h as compared with the extracellular concentration in the cell culture medium. However, there was no interference of nickel(II) with the uptake of cisplatin into V79 cells. Moreover, nickel(II) did not considerably alter the platinum content of the DNA after cisplatin treatment: whereas 2 h treatment of V79 cells with 50 µM cisplatin led to the formation of 1.13 adducts/104 bp, this value was only slightly increased to 1.27 adducts/104 bp in the presence of 100 µM nickel(II). With respect to transplatin, the intracellular platinum content rose with increased transplatin concentrations, but there was no intracellular accumulation of platinum. Again, this uptake was not altered in the presence of nickel(II) (Figure 5B). Interference of nickel(II) with the repair of DNA lesions induced by cisplatin and transplatin Previously, it had been shown that nickel(II) interferes with the repair of UV-induced DNA repair by inhibiting the incision step of NER (6). To find out whether the sensitivity of the cells towards cisplatin and transplatin in the presence of nickel(II) may be due to an interference with DNA repair as well, we investigated the effect of nickel(II) on the repair process of cisplatin- and transplatin-induced DNA lesions by measuring the transient generation of DNA strand breaks determined by alkaline unwinding. Since the number of DNA strand breaks is usually low due to rapid ligation of repair patches, we measured the repair of cisplatin-induced DNA adducts in the presence of 10 mM hydroxyurea. This metabolic inhibitor causes a depletion of nucleotides by inhibiting ribonucleotide reductase; as a consequence, repair patches stay open for a prolonged period of time, thereby increasing the sensitivity of the test system (29). V79 cells were pre-incubated with NiCl2 for 24 h, incubated for 1 h with cisplatin and allowed to repair in the presence of nickel for up to 5 h (Figure 6). Regarding cisplatin alone, low but constant frequencies of DNA strand breaks were measured during the whole repair time, indicating ongoing repair. The presence of nickel(II) caused a time-dependent accumulation of DNA strand breaks, pointing towards an inhibition of the polymerization/ligation 1180

step of the repair process (Figure 6A). As determined 3 h after cisplatin treatment, this effect was dose dependent and concentrations as low as 50 µM nickel(II) were effective (Figure 6B). To elucidate whether nickel(II) additionally interferes with the incision step of cisplatin-induced DNA repair, the total frequency of incisions was determined. DNA strand breaks accumulated in the presence of the inhibitors hydroxyurea and aphidicolin. These inhibitors block DNA synthesis, while incisions at DNA lesions continue, thereby causing an accumulation of DNA strand breaks which correspond to the total frequency of incisions (29). V79 cells were pre-incubated with nickel(II), incubated with cisplatin for 1 h and allowed to repair in the presence of nickel(II) for 3 h (Figure 7). Regarding cisplatin alone, ~0.7 lesions/106 bp accumulated within 3 h. The presence of 25 µM NiCl2 had no effect; at higher concentrations up to 100 µM nickel(II) there was a slight but not significant reduction in incision events. In a next step, we investigated the effect of nickel(II) on the transient generation of DNA strand breaks during the repair of transplatin-induced DNA lesions. Regarding transplatin alone, ~0.1 DNA strand breaks/106 bp were measured after 3 h repair, whereas in the presence of 25–250 µM NiCl2 no occurrence of DNA strand breaks was detected (data not shown). To further investigate whether the observed inhibition of DNA repair may be due to interference with the incision step, the total incision frequency was quantified in the presence or absence of nickel(II) (Figure 8). With respect to transplatin alone, there was a time-dependent accumulation of DNA strand breaks. In contrast, the combined treatment with transplatin and 250 µM nickel(II) caused a decrease in incision events at all time points; this effect was significant at 5 h post-incubation. Hence, nickel(II) caused an inhibition of transplatin-induced DNA repair by interference with the incision step of nucleotide excision repair. Interference of nickel(II) with XPA To find out whether the observed DNA repair inhibition caused by nickel(II) may be due to an interference with DNA damage recognition, we examined the effect of NiCl2 on specific binding of the XPA protein to damaged DNA. This repair protein is an essential part of the NER machinery that enables formation of the human excision nuclease complex (for a


Fig. 6. Effect of nickel(II) on the induction and closure of DNA strand breaks after treatment with cisplatin in Chinese hamster cells. V79 cells were preincubated with 10 mM hydroxyurea for 1 h, treated with 100 µM cisplatin for 1 h and allowed to repair for different time periods in the presence of hydroxyurea. When investigating the effect of nickel(II), the cells were pre-incubated with 250 µM NiCl2 for 20 h before cisplatin treatment in the absence of nickel(II). DNA repair was carried out in the presence of nickel(II). DNA strand breaks were analysed by the alkaline unwinding technique; the number of DNA strand breaks induced by hydroxyurea in the absence or presence of nickel(II) has been subtracted. (A) Effect of 250 µM nickel(II) at different time points after cisplatin treatment; (B) effect of nickel(II) as a function of dose 3 h after cisplatin treatment. Shown are mean values 6 SD from three determinations.

Fig. 7. Effect of nickel(II) on the incision frequency after cisplatin treatment in Chinese hamster cells. V79 cells were pre-incubated with 15 µM aphidicolin and 10 mM hydroxyurea for 1 h, incubated with 100 µM cisplatin for 1 h and allowed to repair for 3 h in the presence of the inhibitors. When investigating the effect of nickel(II), the cells were preincubated with nickel(II) for 20 h before cisplatin treatment in the absence of nickel(II). DNA repair was carried out in the presence of nickel(II) for 3 h. DNA strand breaks were analysed by the alkaline unwinding technique; the number of DNA strand breaks induced by aphidicolin plus hydroxyurea in the absence or presence of nickel(II) has been substracted. Shown are mean values 6 SD of three determinations. The decrease in the incision frequency was not significant as determined by Student’s t-test.

Fig. 8. Effect of nickel(II) on the incision frequency after transplatin treatment in V79 cells. The cells were pre-incubated with 15 µM aphidicolin and 10 mM hydroxyurea for 1 h, incubated with 100 µM transplatin for 1 h and allowed to repair for different time periods up to 5 h in the presence of the inhibitors. When investigating the effect of nickel(II), the cells were pre-incubated with nickel(II) for 20 h before transplatin treatment in the absence of nickel(II). DNA repair was carried out in the presence of nickel(II). DNA strand breaks were analysed by the alkaline unwinding technique; the number of DNA strand breaks induced by hydroxyurea in the absence or presence of nickel(II) has been subtracted. Shown are mean values 6 SD of three determinations. *Significant as determined by Student’s t-test (P , 0.05).

review see ref. 30). Recent studies have shown that XPA preferentially binds to UV-irradiated and cisplatin-damaged DNA as compared with undamaged DNA (19). For assessing metal toxicity, this protein is of special interest, since it contains a zinc finger structure involved in DNA binding (30,31), which may be a sensitive target for toxic metal ions. Therefore, the gel mobility shift assay was used to detect the DNA-binding activity of the XPA protein in the absence or presence of nickel(II) (Figure 9). No non-specific binding of the XPA protein to undamaged DNA was detected, whereas the binding activity to cisplatin-damaged DNA was sufficient to cause a gel mobility shift. While 5 µM NiCl2 had no effect on the XPA–DNA binding activity, concentrations of 10 µM NiCl2 and higher disturbed the binding behaviour (Figure 9A).

Thus, 50 µM NiCl2 led to a 50% reduction in the DNAbinding activity and 75 µM NiCl2 caused nearly complete inhibition (Figure 9B). Discussion The results obtained in the present study demonstrate that nickel(II) increases the sensitivity of V79 Chinese hamster cells to cisplatin, transplatin and mitomycin C. While nickel(II) did not alter drug uptake and DNA adduct formation, it interfered with the repair of cisplatin- and transplatin-induced DNA lesions, providing a plausible explanation for the enhancement of cytotoxicity. Interestingly, different steps of the repair processes were 1181


Fig. 9. Effect of nickel(II) on XPA binding activity to cisplatin-damaged DNA. (A) Aliquots of 500 ng XPA protein were incubated with up to 75 µM NiCl2 where indicated before conduction of the gel mobility shift assay as described in Materials and methods. (B) Quantification of the data after densitometric evaluation. Shown are mean values of two independent experiments; black dots indicate single values obtained in the respective experiments.

affected by nickel(II) in combination with cisplatin and transplatin, respectively. With regard to transplatin, nickel(II) reduced the frequency of incision, which supports results obtained previously, where nickel(II) inhibited the incisions generated after UV irradiation (6). Since transplatin-induced DNA adducts are, like UV-induced DNA photoproducts, repaired via NER (10), the data point towards a general inhibition of this repair system by nickel(II). The diminished incision frequency may be due to the disruption of DNA– protein interactions involved in the recognition of DNA damage. Thus, in the present study we have demonstrated a decreased binding activity of XPA towards a cisplatin-damaged oligonucleotide in the presence of nickel(II). Even though recent experiments have demonstrated that global genome NER is initiated by the XPC–HR23B protein complex, XPA is thought to increase the specificity of damage recognition and discrimination and to organize and correctly orientate the NER machinery around the DNA injury (32). Further support for interference of DNA damage reognition by nickel was obtained in a previous study: by applying a UV-modified synthetic oligonucleotide and HeLa nuclear extracts, we 1182

recently reported that nickel(II) diminished the preferential binding of nuclear proteins to UV-damaged DNA (7). In contrast, with respect to cisplatin we observed only a small, non-significant reduction in the incision frequency in the present study. Instead, DNA strand breaks accumulated after incubation with cisplatin in the presence of nickel(II), indicating interference with a later step of the repair process, like polymerization and/or ligation. One possible explanation for this discrepancy is the contribution of different DNA repair systems during the first hours after drug treatment. Even though the major DNA lesions induced by cisplatin, 1,2d(GpG) and 1,2-d(ApG) intrastrand crosslinks, are in principle targets of NER (12,33), their excision proceeds slowly as compared with other substrates of NER, like cisplatin-induced 1,3-GTG intrastrand crosslinks or DNA lesions induced by 2-acetylaminofluorene, presumably due to different structural alterations of the DNA helix (34). In support of this, investigations applying an in vitro repair system show that 1,2-intrastrand crosslinks do not contribute to the repair synthesis observed on cisplatin-damaged plasmids. Instead, it has been demonstrated that the cisplatin-induced DNA–DNA interstrand crosslink represents a major lesion leading to repair synthesis (35). Concerning the effect of nickel(II) on the repair of cisplatin-induced DNA lesions observed in the present study, the accumulation of DNA strand breaks could therefore result from interactions with crosslink repair, which is also supported by the enhancement of mitomycin C-induced cytotoxicity in the presence of nickel(II). However, if this assumption is correct, why does nickel(II) affect the incision step in NER and a later step in crosslink repair? Even though crosslink repair in mammalian cells is not well understood, experimental evidence suggests that two different repair systems are involved in the removal of DNA–DNA crosslinks, namely NER and recombinational repair. With respect to NER, it has recently been shown that this repair system makes dual incisions 59 of the crosslinked base, thereby removing a damage-free oligonucleotide close to the lesion, which may act as a signal to initiate crosslink removal by recombinational steps (36). However, there also seems to exist a NER-independent crosslink repair system. When using Chinese hamster repairdeficient mutants, they exhibit two distinct levels of sensitivity to DNA–DNA crosslinking agents. NER-deficient cell lines exert only a moderately increased sensitivity, while mutations in ERCC1 and ERCC4 (XPF) provoke an extraordinarily increased cytotoxicity of DNA–DNA crosslinking agents (37). This may be explained by the fact that the latter two proteins are not only part of the NER complex, but, as supported by studies derived with their yeast counterparts RAD1 and RAD10, are also involved in recombinational events (38,39), thus rendering ERCC1 and ERCC4 mutants completely deficient in crosslink repair. With respect to the results obtained in the present study, one possible explanation for the differential effects of nickel(II) on the repair of transplatin and cisplatin-induced DNA lesions could therefore be inhibition of damage recognition in NER, leading to pronounced diminished incision at sites of transplatin-induced DNA lesions, but only to a small decrease in incision at cisplatin-induced DNA–DNA crosslinks, since the latter type of lesion can also be repaired by a NER-independent DNA repair pathway. Instead, nickel(II) inhibits the polymerization/ligation step of crosslink repair, which is in agreement with previous observations according to which nickel(II) also inactivates post-incision steps in NER (5,6).


Concerning possible mechanisms of repair inhibition by nickel(II), competition with essential metal ions may be important. One class of potentially critical targets may be zinc finger structures in DNA-binding motifs of several DNA repair proteins, including XPA. In support of this theory, nickel(II) was able to displace zinc from zinc finger structures of some transcription factors, like the bovine oestrogen receptor (40) and SP1 (41). Alternatively, nickel(II) could displace magnesium(II) from sites essential for DNA–protein or protein–protein interactions, since nickel(II)-induced repair inhibition after UV irradiation was reversible by the addition of magnesium(II) (6,7). What mechanism accounts for inactivation of the XPA protein will be further investigated. Taken together, the results presented in this study show that nickel(II) at low, non-cytotoxic concentrations inhibits NER in general, but also suggest interference with DNA crosslink repair. It has to be kept in mind, however, that nickel(II) could additionally affect the cellular response towards cytostatic drugs on other levels as well, for example by interference with apoptosis and/or cell cycle control; this aspect will be further investigated. Concerning the practical consequences of our findings, two aspects appear to be important. Since NER and crosslink repair are involved in the removal of DNA damage induced by many environmental and food mutagens, simultaneous exposure towards nickel(II) increases the genotoxic risk of these DNA damaging compounds. In combination with cytostatic drugs, however, simultaneous treatment with nickel(II) could offer an interesting approach to reduce the repair capacity of tumour cells, thereby enabling the application of lower concentrations of the cytostatic drug. Acknowledgements The XPA protein was a kind gift from Dr Eker, Rotterdam, Netherlands. This work was supported by the Deutsche Forschungsgemeinschaft, grant no. Ha 2372/1-1.

References 1. IARC (1990) IARC Monographs on the Evaluation of Carcinogenic Risks to Humans, Vol. 49, Nickel and Nickel Compounds. IARC, Lyon. 2. Hartwig,A. and Beyersmann,D. (1989) Enhancement of UV-induced mutagenesis and sister-chromatid exchanges by nickel ions in V79 cells: evidence for inhibition of DNA repair. Mutat. Res., 217, 65–73. 3. Snyder,R.D., Davis,G.F. and Lachmann,P.J. (1989) Inhibition by metals of X-ray and ultraviolet-induced DNA repair in human cells. Biol. Trace Elem. Res., 21, 389–398. 4. Hartwig,A. and Beyersmann,D. (1989) Comutagenicity and inhibition of DNA repair by metal ions in mammalian cells. Biol. Trace Elem. Res., 21, 359–365. 5. Lee-Chen,S.F., Wang,M.C., Yu,C.T., Wu,D.R. and Jan,K.Y. (1993) Nickel chloride inhibits the DNA repair of UV-treated but not methyl methanesulfonate-treated Chinese hamster ovary cells. Biol. Trace Elem. Res., 37, 39–50. 6. Hartwig,A., Mullenders,L.H.F., Schlepegrell,R., Kasten,U. and Beyersmann,D. (1994) Nickel(II) interferes with the incision step in nucleotide excision repair. Cancer Res., 54, 4045–4051. 7. Hartmann,M. and Hartwig,A. (1998) Disturbance of DNA damage recognition after UV-irradiation by nickel(II) and cadmium(II) in mammalian cells. Carcinogenesis, 19, 617–621. 8. Lepre,C.A. and Lippard,S.J. (1990) Interaction of platinum antitumor compounds with DNA. In Eckstein,F. and Lilley,D.M.J. (eds) Nucleic Acids and Molecular Biology. Springer Verlag, Berlin, Germany, Vol. 4, pp. 9–38. 9. Fichtinger-Schepman,A.M., van Dijk-Knijnenburg,H.C., van der VeldeVisser,S.D. Berends,F., Baan,R.A. (1995) Cisplatin– and carboplatin–DNA adducts: is PT–AG the cytotoxic lesion? Carcinogenesis, 16, 2447–2453. 10. Hansson,J. and Wood,R.D. (1989) Repair synthesis by human cell extracts

in DNA damaged by cis- and trans-diamminedichloroplatinum(II). Nucleic Acids Res., 17, 8073–8091. 11. Szymkowski,D.E., Yarema,K., Essigmann,J.M., Lippard,S.J. and Wood,R.D. (1992) An intrastrand d(GpG) platinum crosslink in duplex M13 DNA is refractory to repair by human cell extracts. Proc. Natl Acad. Sci. USA, 89, 10772–10776. 12. Zamble,D.B., Mu,D., Reardon,J.T., Sancar,A. and Lippard,S.J. (1996) Repair of cisplatin–DNA adducts by the mammalian excision nuclease. Biochemistry, 35, 10004–10013. 13. Fraval,H.N.A., Rawlings,C.J. and Roberts,J.J. (1978) Increased sensitivity of UV-repair-deficient human cells to DNA bound platinum products which unlike thymine dimers are not recognized by an endonuclease extracted from Micrococcus luteus. Mutat. Res., 51, 121–132. 14. Plooy,A.C.M., van Dijk,M., Berends,F. and Lohman,P.H.M. (1985) Formation and repair of DNA interstrand cross-links in relation to cytotoxicity and unscheduled DNA synthesis induced in control and mutant human cells treated with cis-diamminedichloroplatinum(II). Cancer Res., 45, 4178–4184. 15. Dijt,F., Fichtinger-Schepman,A.M., Berends,F. and Reedijk,J. (1988) Formation and repair of cisplatin-induced adducts to DNA in cultured normal and repair-deficient human fibroblasts. Cancer Res., 48, 6058–6062. 16. Tomasz,M. (1994) DNA adducts of the mitomycins. In Hemminiki,K., Dipple,A., Shuker,E.E.G., Kadlubar,F.F., Segerback,D. and Bartsch,H. (eds) DNA Adducts: Identification and Biological Significance. IARC, Lyon, pp. 349–357. 17. Ishida,R. and Buchwald,M. (1982) Susceptibility of Fanconi’s anemia lymphoblasts to DNA cross-linking and alkylating agents. Cancer Res., 42, 4000–4006. 18. Sognier,M.A. and Hittelman,W.N. (1983) Loss of repairability of DNA interstrand cross-links in Fanconi’s anemia cells with culture age. Mutat. Res., 108, 383–393. 19. Jones,C.J. and Wood,R.D. (1993) Preferential binding of the xeroderma pigmentosum group A complementing protein to damaged DNA. Biochemistry, 32, 12096–12116. 20. Asahina,H., Kuraoka,I., Shirakawa,M., Morita,E.H., Miura,N., Miyamoto,I., Ohtsuka,E., Okada,Y. and Tanaka,K. (1994) The XPA protein is a zinc metalloprotein with an ability to recognize various kinds of DNA damage. Mutat. Res., 315, 229–237. 21. Kuraoka,I., Morita,E.H., Saijo,M., Matsuda,T., Morikawa,K., Shirakawa,M. and Tanaka,K. (1996) Identification of a damaged-DNA binding domain of the XPA protein. Mutat. Res., 362, 87–95. 22. Moggs,J.G., Yarema,K.J., Essigmann,J.M. and Wood,R.D. (1996) Analysis of incision sites produced by human cell extracts and purified proteins during nucleotide excision repair of a 1,3-intrastrand d(GpTpG)–cisplatin adduct. J. Biol. Chem., 271, 7177–7186. 23. Sehlmeyer,U., Hechtenberg,S., Klyszcz,H. and Beyersmann,D. (1990) Accumulation of chromium in Chinese hamster V79 cells and nuclei. Arch. Toxicol., 64, 506–508. 24. Ahnstro¨m,G. and Erixon,A. (1981) Measurement of strand breaks by alkaline denaturation and hydroxyapatite chromatography. In Friedberg,E.C. and Hanawalt,P.C. (eds) DNA Repair. Marcel Dekker, New York, NY, Vol. 1, pp. 403–418. 25. Hartwig,A., Klyszcz-Nasko,H., Schlepegrell,R. and Beyersmann,D. (1993) Cellular damage by ferric nitrilotriacetate and ferric citrate in V79 cells: interrelationship between lipid peroxidation, DNA strand breaks and sisterchromatid exchanges. Carcinogenesis, 14, 107–112. 26. Hartwig,A., Gro¨blinghoff,U.D., Beyersmann,D., Natarajan,A.T., Filon,R. and Mullenders,L.H.F. (1997) Interaction of arsenic(III) with nucleotide excision repair in UV-irradiated human fibroblasts. Carcinogenesis, 18, 399–405. 27. Garabedian,M.J., LaBaer,J., Wie-Hong,L. and Thomas,J.R. (1993) Analysis of protein–DNA interactions. In Hames,B.D. and Higgins,S.J. (eds) Gene Transcription: A Practical Approach. IRL Press, Oxford, UK. 28. Plooy,A.C.M., van Dijk,M. and Lohman,P.H.M. (1984) Induction and repair of DNA cross-links in Chinese hamster ovary cells treated with various platinum coordination compounds in relation to platinum binding to DNA, cytotoxicity, mutagenicity and antitumor activity. Cancer Res., 44, 2043–2051. 29. Snyder,R.D., Van Houten,B. and Regan,J.D. (1984) The accumulation of DNA strand breaks due to incision; comparative studies with various inhibitors. In Collins,A., Downes,C.S. and Johnson,R.T. (eds) DNA Repair and its Inhibition. IRL, Oxford, UK, pp. 13–32. 30. Cleaver,J.E. and States,J.C. (1997) The DNA damage-recognition problem in human and other eukaryotic cells: the XPA damage binding protein. Biochem. J., 328, 1–12. 31. Tanaka,K., Miura,N., Satokata,I., Miyamoto,I., Yoshida,M.C., Satoh,Y., Kondo,S., Yasui,A., Okayama,N. and Okada,Y. (1990) Analysis of a

1183

I.Krueger, L.H.F.Mullenders and A.Hartwig human DNA excision repair gene involved in group A xeroderma pigmentosum and containing a zinc-finger domain. Nature, 348, 73–76. 32. Sugasawa,K., Ng,J.M.Y., Masutani,C., Iwai,S., van der Spek,P.J., Eker,A.P., Hanaoka,F., Bootsma,D. and Hoeijmakers,J.H. (1998) Xeroderma pigmentosum group C protein complex is the initiator of global genome nucleotide excision repair. Mol. Cell, 2, 223–232. 33. Huang,J.-C., Hsu,D.S., Kazantsev,A. and Sancar,A. (1994) Substrate spectrum of human excinuclease: repair of abasic sites, methylated bases, mismatches and bulky adducts. Proc. Natl Acad. Sci. USA, 91, 12213–12217. 34. Moggs,J.G., Szymkowski,D.E., Yamada,M., Karran,P. and Wood,R.D. (1997) Differential human nucleotide excision repair of paired and mispaired cisplatin–DNA adducts. Nucleic Acids Res., 25, 480–490. 35. Calsou,P., Frit,P. and Salles,B. (1992) Repair synthesis by human cell extracts in cisplatin-damaged DNA is preferentially determined by minor adducts. Nucleic Acids Res., 20, 6363–6368. 36. Bessho,T., Mu,D. and Sancar,A. (1997) Initiation of DNA interstrand cross-link repair in humans: the nucleotide excision repair system makes

1184

dual incisions 59 to the cross-linked base and removes a 22- to 28nucleotide-long damage-free strand. Mol. Cell. Biol., 17, 6822–6830. 37. Hoy,C.A., Thompson,L.H., Mooney,C.L. and Salazar,E.P. (1985) Defective DNA cross-link removal in Chinese hamster cell mutants hypersensitive to bifunctional alkylating agents. Cancer Res., 45, 1737–1743. 38. Schiestl,R.H. and Prakash,S. (1988) RAD1, an excision repair gene of Saccharomyces cerevisiae, is also involved in recombination. Mol. Cell. Biol., 8, 3619–3626. 39. Schiestl,R.H. and Prakash,S. (1990) RAD10, an excision repair gene of Saccharomyces cerevisiae, is involved in the RAD1 pathway of mitotic recombination. Mol. Cell. Biol., 10, 2485–2491. 40. Predki,P.F. and Sarkar,B. (1992) Effect of replacement of ‘zinc finger’ zinc on estrogen receptor DNA interactions. J. Biol. Chem., 267, 5842–5846. 41. Thiesen,H.-J. and Bach,C. (1991) Transition metals modulate DNA protein interactions of SP1 zinc finger domains with its cognate target size. Biochem. Biophys. Res. Commun., 176, 551–557. Received October 9, 1998; revised March 1, 1999; accepted March 16, 1999